Hi Zdenko, Thanks for the response. However, ocrd-testset.zip contains training images and ground truth text without boxes.
True, the images contain a full line of text: [image: alexis_ruhe01_1852_0099_012.png] But there are no box files in the training set. I'd like to confirm if the LSTM training set's xxx.box file is expected contain one box per line (wrapping the entire line) or one box per character in the line... Any insight? On Thursday, September 5, 2024 at 9:15:12 PM UTC+8 zdenop wrote: > have a look at provided example ocrd-testset.zip > <https://github.com/tesseract-ocr/tesstrain/blob/main/ocrd-testset.zip> > > Zdenko > > > ut 3. 9. 2024 o 16:04 'Danny' via tesseract-ocr < > [email protected]> napísal(a): > >> @zdenop wrote: >> | Tesseract LSTM engine (tesseract >=v4) training script is based on >> lines (group of words) >> | Box files reflect that. And yes - box files are important. >> >> Zdenko, does this mean a "box file" for LSTM training should wrap the >> entire text line and NOT the individual characters? >> Which is correct for LSTM training: >> >> A) individual boxes like this, or >> [image: sub_2.png] >> B) One box for entire line: >> [image: sub_2 line.png] >> Thanks. >> >> On Sunday, July 14, 2024 at 9:05:48 PM UTC+8 zdenop wrote: >> >>> Ehm: >>> >>> 1. Tesseract v3 (legacy) engine training is based on characters. >>> 2. Tesseract LSTM engine (tesseract >=v4) training script is based >>> on lines (group of words) >>> >>> Box files reflect that. And yes - box files are important. >>> >>> >>> Zdenko >>> >>> >>> pi 12. 7. 2024 o 14:14 Mateusz Matela <[email protected]> napísal(a): >>> >>>> As an experiment, I run the training on a small sample produced with >>>> text2image. Then I converted the .box files so that each character is >>>> assigned common bounding rectangle from all the characters and run the >>>> training again. The outputs were identical in both cases. Then I removed >>>> the box file and let the training script autogenerate them. In that case >>>> the reported error rates were crazy, like 99% instead of 0.5%. >>>> This suggests that conclusion 3 is correct. >>>> >>>> środa, 10 lipca 2024 o 15:17:07 UTC+2 Mateusz Matela napisał(a): >>>> >>>>> Hi all, >>>>> >>>>> Sorry if double posting, my previous message didn't appear and I don't >>>>> see any info about waiting for acceptance or something. >>>>> I was searching for this topic in this forum and it was mentioned a >>>>> few times, but I couldn't find a clear and definitive explanation. >>>>> >>>>> How does the information put in the .box files affect the training >>>>> process? The file contains coordinates for each character in the txt >>>>> file, >>>>> but the documentation says that since Tesseract 4.0 the model operates on >>>>> the level of whole lines. Some tools like text2image generate the .box >>>>> files with accurate coordinates for each character. When the .box files >>>>> are >>>>> missing the tesstrain Makefile generates them using generate_line_box.py, >>>>> which assigns the same full image area to each character. >>>>> >>>>> I see 3 possible conclusions, which one is closest to the truth? >>>>> >>>>> 1. The .box files do not affect the LSTM training at all and are just >>>>> a leftover from the times of Tesseract 3. In that case, ideally in the >>>>> future they could be completely dropped or only required/generated when >>>>> specifically working with the legacy engine. >>>>> >>>>> 2. There is still a chance that training will work better with exact >>>>> coordinates and the generate_line_box.py is just a cheap workaround that >>>>> could be improved on in the future. >>>>> >>>>> 3. The .box file is still important in case you prefer to define the >>>>> coordinates for the text in the image instead of cropping the image. The >>>>> granularity of the coordinates is not imporant as Tesseract will just >>>>> work >>>>> on a box that encapsulates all of the character boxes. Even if confusing, >>>>> this approach is still better than having a different .box file formats >>>>> for >>>>> LSTM and the legacy engine. >>>>> >>>>> I'll be grateful for any wisdom on this. >>>>> >>>>> Thanks >>>>> Mateusz >>>>> >>>> -- >>>> You received this message because you are subscribed to the Google >>>> Groups "tesseract-ocr" group. >>>> To unsubscribe from this group and stop receiving emails from it, send >>>> an email to [email protected]. >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/tesseract-ocr/b17225d5-2b78-41bd-994f-05305b9a443dn%40googlegroups.com >>>> >>>> <https://groups.google.com/d/msgid/tesseract-ocr/b17225d5-2b78-41bd-994f-05305b9a443dn%40googlegroups.com?utm_medium=email&utm_source=footer> >>>> . >>>> >>> -- >> You received this message because you are subscribed to the Google Groups >> "tesseract-ocr" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> > To view this discussion on the web visit >> https://groups.google.com/d/msgid/tesseract-ocr/ba9b210d-a38e-446d-80e1-4d22b213f210n%40googlegroups.com >> >> <https://groups.google.com/d/msgid/tesseract-ocr/ba9b210d-a38e-446d-80e1-4d22b213f210n%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> > -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/3a502cb4-99fb-4ff3-8aab-dfbe854a1b69n%40googlegroups.com.
